Organization Design

Regretted Attrition and What Actually Makes People Leave

Split your turnover number into the departures you would have paid to prevent and the ones you would not, cost the first group properly, and then act on what the evidence says predicts a resignation rather than on what exit interviews politely report.

  • Advanced
  • 14 min total
  • 14 chapters

What decision this helps you make: Which departures you actually regret, what they cost, where they concentrate, and which single intervention has the best claim on the retention budget you have.

What this topic is

Regretted attrition is the subset of voluntary departures you would have paid to prevent. The headline turnover figure mixes it with the people you were glad to see go and the people you were indifferent about, and because the mixture varies, the blended number moves for reasons that have nothing to do with whether you are losing the people you need. Splitting the number is the first and cheapest analytical move available, and most companies have never made it.

Why it matters

A regretted departure in a mid-level role commonly costs more than half that role's loaded annual cost once recruiting, vacancy, ramp and the team's own time are counted. None of that appears as a line item anywhere. Meanwhile the reasons people give on the way out are systematically unreliable, so companies routinely spend a retention budget on the one thing departing employees find safe to say, which is pay.

Who should learn it

Founders and executives who see a turnover percentage every month and cannot act on it; managers who have lost someone they did not expect to lose; and anyone being asked to approve a retention programme.

What you will understand

  • How to segment a turnover number into regretted, neutral and unregretted, and why the blended figure misleads
  • A defensible build-up of what one regretted departure actually costs
  • Why exit interviews under-report the real reason, and what the turnover research says instead
  • The shocks and the leading indicators that precede a resignation, and what can still be done at each stage

Prerequisites

Common misconception

"People leave for money." Compensation is the most frequently cited reason in exit interviews and one of the weaker predictors in the research, and both facts have the same cause: pay is the safest thing to say. It is impersonal, it does not criticise anyone in the building, and it protects the reference. Meanwhile the strongest available predictors are the ones nobody volunteers on the way out: whether the person has begun looking, whether they believe there is a better alternative available, and whether something specific and jarring happened a few months ago. A retention strategy built from exit-interview transcripts is a strategy built from the one answer designed to end the conversation politely.